DE102004044111B4 - Method and device for coding and reconstructing computer-generated video holograms - Google Patents

Abstract

A method and a device for encoding and reconstructing computer-generated video holograms using a conventional LC display: it provides holographic reconstruction of three-dimensional scenes using electronically controllable pixel in a holographic array (3) with a conventional resolution, and is reasonably free from flickering and cross-talk. Reconstruction is in real time, and for both eyes at the same time, over a large viewing zone. The method takes advantage of an optical focusing means (2) in order to image vertically coherent light emitted by a line light source (1) into viewing windows (8R, 8L) after modulation by the pixel array (3). The holographic reconstruction (11) of the scene is rendered visible from viewing windows (8R, 8L) for both eyes of an observer by way of diffraction at the pixels. According to the invention, the controllable pixels are disposed in vertical pixel columns (15, 16), which encode separate holograms of the same scene for each of the viewer's eyes (R, L), where said holograms are one-dimensional in the vertical direction and horizontally interleaved. An image separation means (7) with separating elements arranged parallel to the pixel columns reveals the respective pixel columns (15, 15' or 16, 16') for one eye and covers them for the other eye.

Description

The invention relates to a method and apparatus for encoding and reconstructing large format computer generated video holograms using minimum resolution displays for a large viewing angle with high spatial image quality. The display includes a panel of controllable matrix pixel arrays that electronically affect the amplitude and / or phase of light.

In the present document, the term pitch denotes the distance between the centers of two adjacent pixels of the display and thus characterizes the display resolution.

Additionally, in the present document, the term encoding refers to the manner of applying control values to the pixel array to holographically reconstruct a three-dimensional scene that is visible through viewer windows.

Suitable displays are currently available as amplitude or phase displays, for example in LCD technology. However, the invention can also be applied to other controllable matrix pixel arrays which use coherent light.

A major problem in encoding and reconstructing large format computer generated video holograms is a sufficiently large viewer area for viewing the reconstruction. Presently available large format displays have pixel arrays with pitches that diffract light only in a very small observer area, so that binocular viewing of a reconstructed three dimensional Scene is not possible. In order to achieve a viewing angle of about 60 °, the pitch of the controllable pixels would have to be about 1 μm. Such high-resolution displays are currently not available. A further disadvantage is that even with the availability of such displays, the required data rates for the calculation, control and transmission of the control data place very high demands on computer technology.

There are several suggestions for solving this problem.

In the arrangement described by Maeno and others in "Electro-holographic Display using 15 Mega Pixels LCD", Advanced 3D Project, 1996, SPIE, Vol. 2652, the problem of enlarging the viewer's range is solved by replacing a conventional LC display low-resolution display of the computer-generated video hologram five special displays with higher resolution can be used. These are placed in the so-called tiling process either directly or by optical imaging together. It is only a horizontal parallax generated on the vertical parallax is omitted. In one example, to display a video hologram, 3200 × 960 pixels are needed to reconstruct a scene of only 50 mm × 150 mm with a depth of 50 mm. The viewing area is 65 mm wide, which corresponds to the distance between the eyes, so that the picture is just visible with both eyes. The required resolution depends on the size of the desired video hologram and observer area. The disadvantage, however, is the required variety of displays, the large lens for reconstruction, the associated large depth and voluminous arrangement and the large computing power required.

Another solution for expanding the viewer area is by Mishina u. a. described in "Viewing zone enlargement method for sampled hologram that uses high-order diffraction", Applied Optics, 2002, Vol. 8. The method reconstructs not only a hologram in the first diffraction order, but also in other orders and combines them to form a common viewer area. The corresponding video holograms for a specific object are shown one after the other on a LC display. By means of a second LC display, which serves as spatial frequency filter, the different diffraction orders are filtered out during the reconstruction. The visible subregions are generated in a time sequential manner and are strung together spatially. The achievable viewer area is still less than 65 mm, so that the reconstructed object is to be considered only with one eye. Another disadvantage is the high computational effort. In addition, extremely fast switchable pixel arrangements are required.

In the time-sequential stringing together of different orders of diffraction, the displays used in addition to the largest possible resolution must also have a high switching speed, if the image should not flicker. This is why mostly binary hologram representations are used, but they produce significant errors due to the binary coding.

A common disadvantage of known holographic methods is the high computational complexity for generating the holograms. From the document WO 2003/021363 A1 is known to dispense with the vertical parallax in the reconstruction and represent this only horizontally. The device includes a computer-generated hologram illumination system which utilizes a line-shaped light source for monochromatic light whose bandwidth is less than 10 nm and that in horizontal Direction coherent but incoherent in the vertical direction.

In the publication WO 2004/044659 A2 the applicant presents a device for reconstructing video holograms with at least one real or virtual point and / or line, sufficiently coherent light source, a lens and matrix-arranged cells with at least one in amplitude and / or phase controllable opening per cell and a viewer plane am Location of the light source image. The device reconstructs a video hologram in a periodicity interval of the Fourier transform in a viewer plane. The reconstructed three-dimensional scene can be viewed with both eyes through viewer windows, with the viewer window extent not larger than the periodicity interval. The scene is visible within a reconstruction cone spanned between the display area and the viewer window, where the scene can be reconstructed in front of, on, or behind the display area. The proposed solution makes it possible to encode the video hologram on a conventional display with currently available resolution of about 3 million pixels with reasonable hardware and computational effort. An extended embodiment of the solution provides for associating with the other eye of the observer a second observer window by connecting a second real or virtual, sufficiently coherent light source at another suitable location in the optical system. In this case, a sequential display due to low frame rate and long switching times leads to a crosstalk between the two eyes.

A line-shaped light source may be, for example, a covered conventional light source whose light radiates through a narrow gap.

For the purposes of this document, a light source can be regarded as sufficiently coherent if its light is spatially coherent so far that it is capable of interfering, so that it is suitable for at least one-dimensional holographic reconstruction with sufficient resolution. These requirements can also meet conventional light sources when they radiate through a sufficiently narrow gap. A line-shaped light source can be considered perpendicular to its length as a point. The light is then coherent in this direction and perpendicular to it but incoherent.

To ensure temporal coherence, the spectrum of the light source must be sufficiently narrowband. The color information can be monochromatically, time-sequentially or spatially divided into spectral components by filter means. The electronically controllable pixels arranged in the form of a matrix modulate the amplitude and / or phase of the interference-capable light. For pixel arrangements which can not directly control the phase of coherent light, the known detour phase method can be used to adjust the phase of the light by means of amplitude settings with several pixels per holographic pixel.

Based on the listed disadvantages of the known solutions, the object of the invention is three-dimensional scenes in the form of computer-generated video holograms by means of electronically controllable matrix pixel arrangements with a minimum possible resolution, as free of flicker and crosstalk and in real time for both eyes simultaneously in a large View viewer area visible, so that the required refresh rate compared to the known time-division multiplexing method is much lower.

To solve the problem, the present patent application is based on a method for encoding and reconstructing video holograms, in which optically focusing means image coherent light of a linear light source through controllable pixels arranged in a matrix into observer windows which are in diffraction order, one in the controllable Pixel holographically encoded scene to reconstruct and make visible through viewer window.

According to the invention, light is used which is coherent only in the vertical direction in order to realize in each case a vertical one-dimensional reconstruction of holograms of the same scene with the controllable pixels for both eyes of a viewer. This is done in such a way that in each case a column group of horizontally nested pixel columns of the matrix-shaped pixel arrangement contains a hologram intended only for one eye. This creates two spatially nested holograms of a scene. That is, all pixel columns of a first column group of the pixel array reconstruct a hologram for a viewer eye, while adjacent pixel columns of the second column group reconstruct the hologram for the other eye. Since the horizontally incoherent light conventionally images the scene, known image separating means having separating elements arranged parallel to the columns can be used to select the two holograms for the left and right eyes, respectively. The image separating means are arranged at a distance from the pixel arrangement and release the column group determined for the corresponding eye and in each case hide the column group for the other eye.

Both eyes see these holograms simultaneously in a separate viewer window. The coherent illumination in the vertical direction of each hologram leads to Reconstruction of the scene in reconstruction spaces, which have the shape of a truncated pyramid, while the incoherent illumination in the horizontal direction allows the use of known image separating means for selecting the holograms for the corresponding eye, to compose a reconstruction of the scene with horizontal parallax when viewed binocularly from both holograms , That is, both holograms differ by a horizontal parallax according to the eye relief. The truncated pyramid shape of the reconstruction spaces arises as a spatial boundary between the edges of the observer windows and the edges of the pixel array, wherein the pixel array can also be encoded so that the reconstruction can also lie in reconstruction spaces that continue behind the pixel array.

As image separating means, a barrier mask may be arranged at a distance from the pixel matrix. Then the separating elements are transparent and non-transparent strips which each release one of the nested groups of columns for the right and left eye of one observer and cover them for the other eye.

The holographically coded columns of a group advantageously reconstruct subpictures corresponding to the eye position from the same scene, which are combined to form an overall reconstruction of the object in two-eyed viewing.

The order of execution of the listed process steps may vary. In particular, the focusing, holographic modulating and separating the holograms can be interchanged.

In order to make the viewer windows available to the viewer in a large area, a position detection system determines the horizontal, vertical and, advantageously, the axial position of the observer's eyes in order to update the position and / or the content of the observer windows as the eye position changes.

With vertical movement of the eye position of the observer, the observer windows are tracked by a corresponding vertical displacement of the light source.

With horizontal movement of the eye position of the observer, the viewer windows are advantageously tracked by horizontally moving the column groups relative to the image separating means. Alternatively, the image separating means, in particular their separating elements relative to the column groups can be moved. The use of the horizontal, linear light source simplifies the tracking of the observer windows.

With axial movement of the eye position of the observer, the distance from the light source and the optically focusing means is changed position-dependent.

The tracking of the viewer window according to the eye position of the viewer in front of the display ensures the visibility of the holographic reconstruction in a wide range with the same constant high quality reproduction

In addition, depending on the eye position by means of software, the coding of the pixel arrangement can be updated so that the holographic reconstruction becomes spatially visible in its spatial position. Alternatively, however, the coding can also be updated such that the reconstructions in their spatial position are correspondingly horizontally and / or vertically displaced for the observer and / or rotated in an angle.

An apparatus for encoding and reconstructing video holograms starts from a coherent light linear light source having vertically focusing means and a pixel array with controllable pixels which modulate the amplitude and / or phase of the coherent light.

According to the invention, the light source is arranged horizontally, so that its light is coherent in the vertical direction and incoherent in the horizontal direction. In controllable pixels of the pixel array, one-dimensional video holograms are encoded in pixel columns such that a first and a second column group for each eye of the observer separately encode a one-dimensional vertically diffracting hologram from the same scene for the respective eye position, with both column groups nested in the horizontal direction , The interleaving of the pixel columns takes place in such a way that image separating means, which lie in the beam path of the light, release the respective column group for the corresponding eye of the observer with separating elements arranged parallel to the columns and obscure for the respective other eye.

As a result, the corresponding hologram for reconstructing the three-dimensional scene is released for each eye in the reconstruction space in front of the viewer windows. Both holograms differ by a horizontal parallax according to the eye distance of the observer. The two holographically coded column groups are two holographic reconstructions of the same scene.

Since the focusing means and the pixel array are to be arranged with a minimum distance to each other, their order is variable. The image separating means may be arranged with their separating elements in the beam path at different locations.

The optically focusing means are advantageously a vertically focusing cylindrical lens, a corresponding Fresnel lens or a lenticular.

Today's transmissive flat panel displays (LCDs) can be used for large-scale reconstructions. For phase encoding you can use the detour phase encoding. The pixel arrangement corresponds to that of a high-resolution transmissive flat panel display. If detour phase encoding or a similar amplitude modulation method of light is used to adjust the phase relationship, the pixels needed for phase adjustment must be perpendicular to each other since the light is not horizontally interfereable. The usual flat display pixel arrangements of adjacent sub-pixels are therefore rotated by 90 °. Pixel arrangements can be used particularly advantageously for the light modulation, the pixels of which directly modulate both the amplitude and the phase of the light, such as the light modulators based on Freedericksz cells.

According to an advantageous embodiment of the invention, a plurality of adjacent pixel columns can also be combined to form a common multiple pixel column. Each multiple pixel column then belongs to one of the two holograms and is made visible as a whole by the image separation means for the corresponding eye. In this case, instead of the simple pixel columns, the multiple pixel columns of both holograms are alternately nested.

Coding the holograms into multiple pixel columns reduces crosstalk. If the viewer moves horizontally, image errors can be effectively suppressed until the observer window is completely tracked. Advantageously, the pixel columns of a multiple pixel column differently modulate the light. For example, of two or more adjacent pixel columns of each multiple pixel columns, only one pixel column is active to modulate the light while the remainder and possibly the next are inactive. That means, they are black keyed.

The range of motion without crosstalk from the hologram for the other eye increases accordingly.

This creates a desired zone between both viewer windows in which no reconstruction is visible. Without this measure, a viewer would perceive a faulty reproduction until the observer windows in this zone were updated. The tracking of the observer windows in the horizontal direction can now be accelerated by the position detection system activating and deactivating different pixel columns during a horizontal head movement within the multiple pixel columns.

The image separating means may be, for example, a barrier mask, a functionally designed lenticular or a prism mask. The image separation elements have an approximately twice as large pitch as the associated column groups of the pixel arrangement in an advantageous embodiment.

For error-free viewing of the holographic reconstruction, a viewer window size is required which corresponds at least to the size of the eye pupil. However, this would place a very high demand on the accuracy and speed of tracking, which would be practically unrealizable, so that the viewer window must be larger in practice. However, the height of the observer window can not exceed the extent of the periodicity interval. It is also necessary to set the viewing window in their width. This is done starting from the width of the pixel columns by matching with the dimensions of the image separating means, in particular the distance and the width of the separating elements.

Both viewer windows are arranged with their centers approximately at eye relief.

A significant advantage of the invention is that instead of a time-multiplex operation of the display with its double frame rate, a space-multiplex operation is realized. At the same time, two spatially interleaved holographic reconstructions of a three-dimensional scene are coded with a single pixel arrangement, which a viewer sees separately as a result of image separation. This allows halving the frame rate.

In addition, the use of one-dimensional, vertically reconstructive holograms in combination with the horizontal incoherent scene mapping significantly reduces the computational overhead of providing the data for encoding. The requirements for the resolution of the pixel arrangement in the horizontal direction are not critical, so that with little effort large-format video holograms can be reconstructed over a large area of the changing eye position of the viewer.

The line-shaped light source used allows a continuous reconstruction in comparison to point-shaped light sources during lateral observer movement.

Another advantage is that for producing the coherent light instead of a laser, for example, a conventional white light source with a slit mask can be used.

The method and the device according to the invention are described in more detail below in an exemplary embodiment. The principle of the invention will be explained with reference to a holographic reconstruction with monochromatic light. However, the object of the invention is also applicable to a color holographic reconstruction. In such a case, the controllable pixels of each pixel column would represent the basic colors needed for color representation in space or time division.

The drawings show:

1 a reconstruction of a 3D scene with two beams between a viewer plane and image separating means, top view

2 a section of the device according to the invention in a three-dimensional representation

3 a section of the device according to the invention in plan view

4 a visible to the viewer reconstruction of the scene in plan view.

In the example shown, the information about a three-dimensional scene is encoded into transmissive pixel arrangements with pixels arranged in a matrix, which are controlled according to the coding, for example by the detour-phase method. However, with regard to the implementation of the pixel arrangement, the implementation of the basic idea of the invention is not limited to the described pixel arrangement. Both transflective and reflective pixel arrangements can be used and / or those which directly modulate the phase of the light waves, such as the Freedericksz cells.

1 shows a focusing means 2 in the form of a horizontally arranged cylindrical lens, which is a linear light source 1 into a viewer level 5 maps. When encoding by the detour phase method, holographic reconstruction uses a higher diffraction order in a periodicity interval.

According to the invention, a horizontally arranged line-shaped light source 1 used as a result of their linear expansion in the vertical direction generates spatially coherent and incoherent in the horizontal direction light. This penetrates transmissive pixels of a matrix-shaped, controllable pixel arrangement 3 , In this case, the holographically coded and columnwise arranged video holograms reconstruct only in the vertical direction, so that in the observer plane through a viewer window, the reconstruction of the holograms in a diffraction order can be considered. In the horizontal direction, due to the incoherence of the light, only one projection of the light source takes place 1 through the image separating agents. 1 shows the image separating means 7 and a ray of light 9 for the right eye R and a ray bundle 10 for the left eye L.

2 shows the device according to the invention in a spatial representation and 3 a section in plan view. The reconstructions of the three-dimensional scene each arise in a truncated pyramid 12 _r , 12 _l between the edges of the pixel array 3 and the viewer windows 8r . 8l , The holographic reconstruction 11 The scene may be in front, on or behind the pixel layout 3 lie. In controllable pixels of the pixel array 3 are one-dimensional video holograms in pixel columns 15 . 15 ' , to 16 '' encoded so that the column group 15 . 15 ' 15 '' for the right eye R of the viewer and the column group 16 . 16 ' . 16 '' for the left eye L each reconstructed a one-dimensional hologram of the same scene, in the example shown, both column groups are alternately nested in the horizontal direction. The nesting of pixel columns 15 to 16 '' takes place in such a way that image separating means 7 , which lie in the beam path of the light, with parallel to the pixel columns 15 to 16 '' arranged separating elements 17 . 18 . 19 the respective column group 15 . 15 ' . 15 '' or 16 . 16 ' . 16 '' for the corresponding eye of the observer and cover for the other eye.

For example, a barrier mask with vertical, alternately transparent or non-transparent strips serves to separate the spatially interleaved holograms 17 . 18 . 19 which is in front of or behind the pixel array 3 is arranged. The transparent stripes give the first column group 15 . 15 ' . 15 '' for the right eye R free. At the same time, this is obscured by the nontransparent strips for the left eye L. The viewer's gaze from the tracked, virtual viewer windows 8r . 8l in the observer level 5 to the corresponding pixel columns 15 and 16 show in 2 the arrows 20 and 21 using the example of a single column 15 . 16 each column group. In reality, several hundreds of pixel columns or pixel column groups are each at the construction of the interleaved holographic reconstructions of the three-dimensional scene 11 involved. Just as many image separation elements must contain the image separating agents.

The distance of the mask of the image separating agent 7 from the pixel layout 3 ( 4 ) and the strip spacing of the preferably uniformly wide strips are selected so that only the video holograms intended for the respective eye are released for the viewer.

An unillustrated position detection system can track the position of observer windows.

The tracking with horizontal change of the eye position can be achieved by moving the separating elements of the image separating means 7 be achieved. This can be done electronically if, for example, a further transmissive pixel arrangement with likewise controllable openings is used for this purpose.

Instead, you can also move the column groups horizontally ( 15 ... 16 ' ) compared to the image separating agents ( 7 ) the viewer window ( 8r . 8l ) track. This can advantageously be carried out electronically with the aid of the mentioned multiple pixel columns. Whenever only one pixel column of the multiple pixel columns is active to modulate the light, horizontal tracking can be assisted by switching to a different pixel column within a multiple pixel column.

Both horizontal shifts can also be performed simultaneously. In the vertical direction, a tracking by vertical displacement of the linear light source 1 be made. An axial position change can be made by changing the distance from the light source 1 and the cylindrical lens 2 compensate

In addition, not only the observer windows are tracked during a movement of the viewer. Also, the contents of the holograms can be recalculated and coded according to the change in perspective.

The peculiarity of the solution according to the invention is that a corresponding video hologram is calculated, coded and reconstructed for each eye R, L of the viewer at the same time, and this is coded and reconstructed with a single pixel arrangement 3 spatially nested, allowing separate holographic reconstructions with the image separating means 7 for the two eyes in each viewer window 8r . 8l can be made visible separately.

Both video holograms have a horizontal parallax with respect to each other that corresponds to the interpupillary distance. Thus, the complete spatial visual impression is guaranteed.

The parallel holographic reconstruction of the scene for both eyes of the viewer ensures a natural vision, so that with the right focus and the accommodation and the convergence of the eyes of the beholder are given to any scene points.

The extent of the observer window lies in a vertical direction within a diffraction order and should not be greater than the periodicity interval in the reconstruction of the video holograms. Otherwise, the viewer sees a superposition of the reconstruction of two adjacent diffraction orders. The size of the windows must also be adapted to the accuracy of the positioning and the speed of the tracking.

By a specific limitation of the observer window during the coding, according to the one from the document WO 2003/021363 A1 known solution, for example, to 10 mm in the vertical direction, reduce the required display resolution and the amount of data to be processed and possibly transferred by at least a factor of 100.

By the present invention, the use of currently available controllable matrices for hologram imaging, e.g. B. in the form of LC flat panel displays, only possible.

In a color coding, the alternately arranged RGB sub-pixels of the three primary colors each reconstruct three individual sub-holograms of a primary color, which form a complete color reconstruction.

4 shows a device according to the invention with all superimposed beams 13 for the right eye and rays 14 for the left eye.

The invention is suitable for entertainment as well as television, multimedia, gaming and mobile terminal applications, as well as commercial applications such as spatial design design, medical and military technology, and many other convenient display techniques.

Claims (21)

Method for coding and reconstructing video holograms, in which optically vertically focusing means ( 2 ) coherent light of a linear light source ( 1 ) after a modulation by matrix-arranged, electronically controllable pixels in a viewer plane ( 5 ) with observer windows ( 8r . 8l ) to diffract by the Pixels to holographically reconstruct a three-dimensional scene within a diffraction order, and to reconstruct it for both eyes (R, L) of an observer from the observer's windows ( 8r . 8l ), characterized in that - the linear light source ( 1 ) radiates vertically coherent light and from the vertically focusing means ( 2 ) for the considered diffraction order at the level of observers ( 5 ) - that controllable pixels within pixel columns ( 15 - 16 '' ) encode a separate, vertically one-dimensional hologram of the same scene for both viewer eyes (R, L), the pixels thus being divided into separate column groups ( 15 . 15 ' . 15 '' or 16 . 16 ' . 16 '' ) are arranged to horizontally nested one-dimensional holograms, and - that image separating means ( 7 ) with parallel to the pixel columns ( 15 - 16 '' ) arranged separating elements the respective column group ( 15 . 15 ' . 15 '' or 16 . 16 ' . 16 '' ) for the corresponding observer eye and obscure for the other eye. Method according to Claim 1, in which the holograms for both eyes (R, L) of the observer encode a horizontal parallax corresponding to the interpupillary distance. The method of claim 1, wherein local changes in eye position in space are detected by a position detection system. Method according to Claims 1 and 3, in which the holographic coding of the pixels in the pixel columns ( 15 - 16 '' ) is updated accordingly when the eye position changes. Method according to Claims 1 and 4, in which, when the eye position is changed, the reconstructed scene ( 11 ) is encoded so that it is visible in its horizontal and / or vertical and / or axial position for the viewer stationary. Method according to Claims 1 and 4, in which the reconstructed scene ( 11 ) is encoded so that it is horizontally and / or vertically shifted and / or rotated in an angle as a function of the change in the eye position of the observer in its horizontal, vertical and axial position. Method according to Claims 1 and 3, in which, when the eye position is changed vertically, vertical displacement of the linear light source ( 1 ) the viewer window ( 8r . 8l ). Method according to Claims 1 and 3, in which, when the eye position is changed horizontally, a horizontal shifting of the column groups ( 15 - 16 '' ) compared to the image separating agents ( 7 ) the viewer window ( 8r . 8l ). Method according to Claims 1 and 3, in which, when the eye position is changed horizontally, a horizontal displacement of the separating elements of the image separating means ( 7 ) against the pixel columns ( 15 - 16 '' ) the viewer window ( 8r . 8l ). Method according to Claims 1 and 3, in which, when the eye position is changed in the light direction, the distance from the light source ( 1 ) and the optically focusing agents ( 2 ) is adjusted depending on the position. Apparatus adapted to carry out a method according to claim 1, comprising - a linear light source coherently emitting in one direction ( 1 ) And optically focusing agents ( 2 ) after a modulation of the light of the light source ( 1 ) by matrix-arranged, controllable pixels in truncated pyramidal reconstruction spaces with observer windows ( 8r . 8l ) to holographically reconstruct a scene, characterized in that - the line-shaped light source ( 1 ) is horizontally arranged so that its light is coherent in the vertical direction, that the controllable pixels within pixel columns ( 15 - 16 '' ) each with a column group ( 15 . 15 ' and 16 . 16 ' ) for each eye (R, L) of a viewer are encoded as one-dimensional, vertically-coded holograms corresponding to the perspective views of the same scene, both column groups ( 15 . 15 ' and 16 . 16 ' ) are nested in the horizontal direction and - that image separating means ( 7 ) with parallel to the pixel columns ( 15 - 16 '' ) arranged separating elements, which the respective column group ( 15 . 15 ' or 16 . 16 ' ) for the corresponding eye of the observer and cover for the other eye. Apparatus according to claim 11, wherein to reduce the crosstalk between the reconstructions of the holograms and / or adjusting the observer windows, a plurality of adjacent pixel columns each form a common multiple pixel column, each multi-pixel column being treated as a whole with the image separating means. 7 ) is selectable for the corresponding eye. The apparatus of claim 12, wherein the pixel columns of a multiple pixel column differently modulate the light. Apparatus according to claim 11, wherein the pixels are designed to be transmissive and / or reflective, and modulate the coherent light in its amplitude and / or in its optical phase. Device according to Claim 11, in which the optically focusing means ( 2 ) are a vertically focusing cylindrical lens, a lenticular or a Fresnel lens. Device according to Claim 11, in which the image separating means ( 7 ) are a barrier mask, a lenticular or a prism mask. Device according to Claim 11, in which the image separating means ( 7 ) are a barrier mask with transparent stripes, which release the respective column group for the corresponding eye of the observer and non-transparent stripes, which obscure the other column group for this eye. Device according to Claim 16, in which the pitch of the separating elements of the barrier mask ( 7 ), the prism mask or the lenticular corresponds to the double horizontal pitch of the associated column group of the controllable pixel array. Device according to Claim 11, in which the controllable pixels are arranged so that, for a detour-phase coding, the pixels or sub-pixels required for phase adjustment are perpendicular to one another for color representation. Apparatus according to claim 12, wherein the extent of the observer windows ( 8r . 8l ) has at least the size of an eye pupil, in the horizontal direction for each eye, by the width of the separating elements of the image separating means ( 7 ) and in which both viewer windows ( 8r . 8l ) are arranged with their centers at eye relief. Apparatus according to claim 12, wherein the vertical extent of the observer windows ( 8r . 8l ) for both eyes (R, L) at most the extension of the periodicity interval of a diffraction order ( 6 ) and at least the size of an eye pupil.

Images